WO2023188995A1 - Vehicle and server - Google Patents

Abstract

In the present invention, a digital twin system comprises a vehicle and a server that generates a digital twin of the vehicle, and the digital twin system realizes a reduction in communication volume between the vehicle and the server without reducing the accuracy of the digital twin. A vehicle (101) comprises: a plurality of sensors (S); an acquisition unit (11) that acquires detection data from each of the plurality of sensors; and a communication unit (2) that transmits at least a portion of the acquired detection data to a server (102) which predicts the future state of the vehicle through input of the detection data, and that receives action instructions based on the prediction from the server. Upon transmitting the detection data, the communication unit transmits the subsequent detection data on the basis of determination results indicating whether the subsequent detection data can be transmitted.

Description

Vehicles and servers

The present invention relates to a vehicle and a server.

Technology related to a digital twin system that reproduces the state of a vehicle in a virtual space based on detection data output by various sensors included in the vehicle is conventionally known. For example, Patent Document 1 describes a step of generating a digital twin of a vehicle, and a step of receiving digital data recorded by sensors and describing the state of the vehicle existing in the real world and the behavior of the vehicle traveling in the real world. , a method is described that includes updating a digital twin of a vehicle based on digital data so that the digital twin is consistent with state and behavior.

JP2020-013557A

By the way, in the digital twin system, the more accurate the state of the vehicle to be reproduced, the more real-time information about the state of the vehicle required by the server that generates the virtual space. However, with the above-mentioned conventional technology, as the reproduction accuracy increases, the amount of communication between the vehicle and the server increases, and there is a possibility that sufficient data cannot be sent to the server. On the other hand, if an attempt is made to suppress the amount of communication between the vehicle and the server, the accuracy of the vehicle state reproduced in the virtual space will decrease.

In order to solve the above problems, a vehicle according to one aspect of the present invention includes a plurality of sensors, an acquisition unit that acquires detection data from the plurality of sensors, and at least one of the detection data acquired by the acquisition unit. a communication unit that transmits the detection data to a server that predicts the future state of the vehicle using a learned model constructed by machine learning, and receives behavior instructions based on the prediction from the server; After transmitting the detection data, the communication unit transmits the next detection data based on a determination result indicating whether or not the next detection data can be transmitted.

In order to solve the above-mentioned problem, a server according to another aspect of the present invention provides a server that includes a plurality of sensors and an acquisition unit that acquires detection data from the plurality of sensors. A communication unit that receives the output detection data; and a communication unit that predicts the future state of the vehicle based on the detection data received by the communication unit using a learned model constructed by machine learning, and uses the prediction result as prediction data. and an execution unit that performs a simulation regarding the traffic situation based on the received detection data and traffic data of traffic participants, and the communication unit, after receiving the detection data, The next detection data is received according to a result of determining whether or not the vehicle can transmit the next detection data.

FIG. 1 is a block diagram showing a digital twin system configured by a vehicle and a server according to an embodiment of one aspect of the present invention. FIG. 2 is a block diagram showing the functional configuration of a vehicle included in the system. FIG. 2 is a perspective view showing the vehicle and sensors included in the vehicle. FIG. 2 is a block diagram showing the functional configuration of a server according to an embodiment. FIG. 3 is a diagram showing a trained model used for prediction by the system. It is a time chart showing the operation of the digital twin system according to the embodiment. FIG. 3 is a diagram illustrating some of the effects of the present invention.

<Embodiment>
First, embodiments of the present invention will be described in detail. FIG. 1 is a block diagram showing a digital twin system 100. FIG. 2 is a block diagram showing the functional configuration of vehicle 101. FIG. 3 is a diagram showing a trained model that the system 100 uses for prediction. FIG. 4 is a perspective view showing the vehicle 101 and sensors included in the vehicle 101. FIG. 5 is a block diagram showing the functional configuration of the server 102.

[Configuration of digital twin system 100]
The digital twin system 100 reproduces a digital twin indicating the state of the vehicle 101 in a virtual space. As shown in FIG. 1, the digital twin system 100 includes a plurality of servers 101 and a server 102. These are connected to each other via a communication network N.

[Vehicle 101]
As shown in FIG. 2, the vehicle 101 includes a plurality of types of sensors S, a vehicle-side control section 1, a vehicle-side communication section 2 (communication section), and a display section 3.

(Vehicle side communication section 2)
The vehicle-side communication unit 2 is connected to other devices (for example, a server 102 described later, a terminal device owned or attached to a traffic participant other than the vehicle 101 (pedestrian, person riding a bicycle), a drone, installed on the road) various data, signals, etc. are transmitted and received by wire or wirelessly with the road camera C installed on the road, the road sensor S9 installed on the road, etc. The vehicle-side communication unit 2 according to this embodiment is composed of a communication module.

(Display section 3)
The display unit 3 is provided at a position in the vehicle 101 that is visible to the driver. Further, the display unit 3 displays a screen based on a signal from the vehicle-side control unit 1.

(Vehicle side control unit 1)
As shown in FIG. 2, the vehicle-side control unit 1 includes an acquisition unit 11, a determination unit 12, a first transmission control unit 13, a first reception control unit 14, a determination unit 15, and a second transmission control unit. section 16 and an output control section 17.

- Acquisition unit 11
The acquisition unit 11 according to the present embodiment acquires detection data D1 from a plurality of types of sensors S, respectively. Further, the acquisition unit 11 according to the present embodiment acquires detection data D1 from each sensor S via an input IF (not shown) to which each sensor S is connected. Further, the acquisition unit 11 according to the present embodiment acquires detection data D1 from other devices (road camera C, road sensor S9, etc.). The acquisition unit 11 according to the present embodiment acquires detection data D1 from another device via the vehicle-side communication unit 2. Further, the acquisition unit 11 according to the present embodiment repeatedly acquires various detection data D1 every time a predetermined time (for example, within a range of 50 to 150 msec) has elapsed. The predetermined time can be set to any length.

The sensors S included in the vehicle 101 include, for example, a position (distance/angle) sensor S1, a speed sensor S2, an acceleration sensor S3, a pressure sensor S4, a temperature sensor S5, a force (torque) sensor S6, and a flow rate sensor as shown in FIG. The gas sensor S7 includes at least one of a gas sensor S7 and a gas sensor S8.

The position (distance/angle) sensor S1 includes a sensor S11 that detects the distance from an object in front, a sensor S12 that detects the distance from an object in the rear, a sensor S13 that detects the rotation angle of the steering wheel, and a throttle valve. The sensor S14 detects the inclination angle of the accelerator pedal, the sensor S15 detects the inclination angle of the accelerator pedal, and the sensor S16 detects the inclination angle of the brake pedal.

The speed sensor S2 includes a sensor S21 that detects the rotational speed of the wheel, a sensor S22 that detects the speed of the crankshaft, a sensor S23 that detects the speed of the camshaft, and a sensor S24 that detects the injection speed of the injection pump in the diesel engine. Contains at least one of them.

The acceleration sensor S3 detects acceleration (impact) acting on the vehicle body.

The pressure sensors S4 include a sensor S41 for detecting tire air pressure, a sensor S42 for detecting brake pressure, a sensor S43 for detecting hydraulic reservoir pressure in power steering, a sensor S44 for detecting suction pressure, a sensor S45 for detecting filling pressure, and a sensor S45 for detecting fuel pressure. , a sensor S47 that detects refrigerant pressure in an air conditioner, and a sensor S48 that detects modulated pressure in an automatic transmission.

The temperature sensor S5 includes a sensor S51 for detecting tire temperature, a sensor S52 for detecting supply air temperature, a sensor S53 for detecting ambient temperature, a sensor S54 for detecting internal temperature, a sensor S55 for detecting evaporator temperature in air conditioning, and a sensor S55 for detecting coolant temperature. and a sensor S57 that detects engine oil temperature.

The force (torque) sensor S6 includes a sensor S61 that detects the force of pressing the pedal, a sensor S62 that detects the weight of the occupant, a sensor S63 that detects the torque acting on the drive shaft, and a sensor S63 that detects the torque acting on the steering wheel. Contains at least one of S64.

The flow meter S7 includes at least one of a sensor S71 that detects the flow rate of fuel and the amount of fuel supplied to the engine, and a sensor S72 that detects the amount of air sucked into the engine.

The gas sensor S8 includes at least one of a sensor S81 that detects the composition of exhaust gas and a sensor S82 that detects harmful substances contained in the supplied air.

Note that the various sensors S described above are known ones, as disclosed, for example, on the following Web page.
・Vehicle functions and types
https://innovationdiscoveries.space/vehicle-sensors-functions-and-types/
・Automotive sensors: the design engineer's guide
https://www.avnet.com/wps/portal/abacus/solutions/markets/automotive-and-transportation/automotive/communications-and-connectivity/automotive-sensors/

Note that the acquisition unit 11 may be configured to acquire the detection data D1 from a plurality of vehicles 101, respectively. Further, the acquisition unit 11 may acquire the detection data D1 (once stored in the storage device) from the vehicle 101 via a storage device (not shown). Further, the acquisition unit 11 may acquire the detection data D1 not through the vehicle-side communication unit 2 but through a recording medium or the like. Further, the acquisition unit 11 may acquire the detection data D1 from a traffic participant other than the vehicle 101, a drone, or the like. Furthermore, as shown in FIG. 2, the acquisition unit 11 may acquire the detection data D1 from at least one of the road camera C and the road sensor S9 via the vehicle-side communication unit 2. Further, the acquisition unit 11 may acquire route information set in the navigation system of the vehicle 101 or the terminal device as the detection data D1. Further, the acquisition unit 11 may be configured to acquire an image taken by a drive recorder or a rear camera included in the vehicle 101 as the detection data D1. Further, the acquisition unit 11 may be configured to acquire information (driver's drowsiness, etc.) determined by the vehicle 101 based on the detection data D1 as the detection data D1.

The determination unit 12 determines whether or not the detection data D1 can be transmitted. By providing the vehicle 101 with such a determination unit 12, determination can be made quickly. The determination unit 12 according to the present embodiment determines whether or not the future predicted by the server 102 (described later) will be reached, as whether or not information can be transmitted. "Reaching the future" may refer to the time just before the future (for example, several seconds ago), or may refer to actually reaching the future. The determining unit 12 may include a timer function, and may be configured to use this function to determine whether or not the above-mentioned future will be reached. Further, if the server 102 (to be described later) can identify the usage information using information from other than the vehicle 101, the determination unit 12 determines that the future has been reached based on the instruction from the server 102, and transmits the usage information. Also good. Cases in which usage information can be specified using information from sources other than the vehicle 101 include, for example, cases in which the server 102 can directly acquire detection data D1 such as position information and speed of the vehicle 101 from the road camera C, road sensor S9, etc. . Further, the determination unit 12 according to the present embodiment repeatedly performs such determination.

・First transmission control unit 13
The first transmission control unit 13 transmits at least a portion of the detection data D1 acquired by the acquisition unit 11 (for example, only the detection data D1 determined to be transmittable by the determination unit 12) to the server 102. Further, the first transmission control section 13 controls the vehicle-side communication section 2 based on the determination result of the determination section 12. Thereby, the vehicle-side communication unit 2 wirelessly transmits at least a portion of the detection data D1 to the server 102 when the determination unit 12 determines that the predicted future will be reached.

Furthermore, as described above, the determination unit 12 according to the present embodiment repeatedly performs determination. Therefore, after transmitting the detection data D1, the vehicle-side communication unit 2 according to the present embodiment wirelessly transmits the next detection data D1 based on the determination result indicating whether or not the next detection data D1 can be transmitted. I will do it.

Furthermore, as described above, the determination unit 12 according to the present embodiment determines whether or not the predicted future will be reached, as whether or not it is possible to transmit the information. Therefore, the vehicle-side communication unit 2 according to the present embodiment wirelessly transmits the detection data D1 to the server 102 when the determination unit 12 determines that the predicted future will be reached. By doing so, there is no need to transmit data to the server 102 until the predicted future is reached, so the amount of communication between the vehicle 101 and the server 102 can be reduced. Note that when the server 102 can directly acquire the detection data D1 from the road camera C, the road sensor S9, etc., the first transmission control unit 13 may control the vehicle-side communication unit 2 to stop transmitting the detection data D1. good.

・First reception control unit 14
The first reception control unit 14 controls the vehicle-side communication unit 2 to receive behavior instructions based on the prediction from the server 102. The behavior instruction indicates, for example, the future position, speed, acceleration, etc. of the vehicle 101. Further, the first reception control unit 14 controls the vehicle-side communication unit 2 to receive the simulation result (digital twin) from the server 102. The digital twin mentioned here includes both the meaning of performing a simulation on the server 102 and the fact that the simulation and the simulation results are fed back to the vehicle 101.

・Judgment part 15
The determining unit 15 determines whether or not it is possible to drive based on the behavior instruction received by the vehicle-side communication unit 2. Specifically, the determination unit 15 compares the content of the behavior instruction with the current state of the vehicle 101 based on the immediately acquired detection data D1, and determines whether the behavior instruction is met when the future arrives. Judge whether the robot can reach the desired position, reach the speed specified by the behavior instructions (lower the speed), etc.

-Second transmission control section 16
The second transmission control unit 16 controls the vehicle-side communication unit 2 based on the determination result of the determination unit 15. Thereby, the vehicle-side communication unit 2 transmits the detection data D1 to the server 102 when the determination unit 15 determines that driving based on the behavior instruction is not possible.

・Output control section 17
The output control unit 17 outputs the received behavior instruction. The output control unit 17 according to this embodiment controls the display unit 3 to display behavior instructions. Further, the output control unit 17 controls the display unit 3 to display the simulation results received from the server 102. Note that the output control unit 17 may be configured to control a speaker (not shown) to output audio.

- Others Note that the vehicle 101 may include a driving control section instead of the display section 3 and the output control section 17 or in addition to the display section 3 and the output control section 17. The driving control unit automatically controls the operation of at least a portion of the vehicle 101 based on the simulation results received from the server 102. In this way, traffic information is provided to the driving control unit more quickly. For this reason, the operation control unit can operate with higher safety. Further, the vehicle-side control unit 1 may include a comparison unit that compares the future state based on the behavior instruction received from the server 102 and the current state based on the detection data D1 acquired by the acquisition unit 11. . In that case, the determination unit 12 of the vehicle 101 determines whether an error greater than a threshold value has occurred between the future state based on the behavior instruction received from the server 102 and the current state based on the detection data D1 acquired by the acquisition unit 11. It may be configured to determine whether or not the usage information can be transmitted.

[Server 102]
The server 102 predicts the future state of the vehicle 101 by inputting the detection data D1. As shown in FIG. 4, the server 102 includes a server-side control section 4, a server-side communication section 5, and a storage section 6.

(Server side communication department 5)
The server-side communication unit 5 exchanges various data and various signals with other devices (for example, the vehicle 101, a terminal device owned by a traffic participant other than the vehicle 101, a drone, a road camera C, a road sensor S9, etc.). etc., can be sent and received by wire or wirelessly. The server-side communication unit 5 according to this embodiment is composed of a communication module.

(Storage unit 6)
The storage unit 6 stores a learned model 61. The trained model 61 according to this embodiment is configured to generate prediction data D2 based on a plurality of detection data D1. As shown in FIG. 5, the detection data D1 includes advance information D11 that is set before driving, event information D12 that occurs when an event occurs, and update information D13 that is periodically updated. The prior information D11 includes, for example, route information of a navigation system. The event information D12 includes, for example, braking information regarding braking of the vehicle 101. The braking information includes, for example, steering information indicating that the steering wheel has been operated, brake information indicating that the brake has been depressed, and turn signal information indicating that the blinker has been blinked. The update information D13 includes, for example, the position of the vehicle 101, the speed of the vehicle 101, the acceleration/deceleration of the vehicle 101, and the like. The update information D13 may be from the sensor S of the vehicle 101, or may be from the road camera C or the road sensor S9. The prediction data D2 includes, for example, future traffic information (eg, position, etc.) of the vehicle 101.

The learned model 61 according to the present embodiment includes previously obtained detection data D1 and traffic information obtained when the detection data D1 was obtained or in a situation similar to when the detection data D1 was obtained. It was constructed using machine learning (for example, deep learning) using the set of as training data. Note that the learned model 61 may be configured to generate other prediction data D2 that is a candidate, different from the prediction data D2.

(Server side control unit 4)
The server-side control unit 4 includes a second reception control unit 41, a prediction unit 42, an execution unit 43, and a third transmission control unit 44.

-Second reception control section 41
The second reception control unit 41 controls the server side communication unit 5 so that the server side communication unit 5 receives the detection data D1 from the vehicle 101. Further, the second reception control unit 41 according to the present embodiment controls the server side communication unit 5 to receive the detection data D1 from the road camera C and the road sensor S9. As described above, after wirelessly transmitting the detection data D1, the vehicle-side communication unit 2 of the vehicle 101 according to the present embodiment wirelessly transmits the next detection data D1 according to the determination result that the next detection data D1 is to be transmitted. Send by. Therefore, after receiving the detection data D1, the second reception control unit 41 causes the vehicle 101 to wirelessly receive the next detection data D1 transmitted according to the determination result that the vehicle 101 transmits the next detection data D1. Controls the communication section 2. Therefore, after receiving the detection data D1, the server side communication unit 5 receives the next detection data D1.

・Prediction unit 42
The prediction unit 42 uses the learned model 61 to predict the future state of the vehicle 101 based on the detection data D1, and outputs the prediction result as prediction data D2. The future refers to, for example, the time after the first predetermined time from the time when the server 102 acquired the detection data D1. The first predetermined time can be set to any length that exceeds the acquisition cycle of the detection data D1 (for example, within a range of 500 to 700 msec). The first predetermined time may include the time from when the server 102 transmits the prediction data D2 until the vehicle 101 receives it. The prediction unit 42 uses the learned model 61 stored in the storage unit 6 to predict the future. The prediction data D2 includes, for example, the future position of the vehicle. Note that the prediction unit 42 may output prediction data D2 with a temporal range instead of prediction data D2 indicating the state at a certain moment in the future. Further, the prediction unit 42 may request the vehicle 101 to transmit the detection data D1 for a new prediction immediately before the predicted future arrives. Further, the prediction unit 42 may be configured to output prediction data D2 having a spatiotemporal range as the prediction data D2 indicating the state at a certain moment in the future.

・Execution unit 43
The execution unit 43 performs a simulation regarding the traffic situation based on the detection data D1 and traffic data of traffic participants around the vehicle 101, and generates a behavior instruction based on the simulation result. The traffic data is, for example, various data transmitted from a terminal device owned or attached to a traffic participant, a drone, a road camera C installed on the road, a road sensor S9 installed on the road, and the like. Note that the simulation results may include not only information regarding the vehicle 101 but also information regarding the surrounding environment of the vehicle 101 (for example, information on other traffic participants, etc.). Furthermore, if the detection data D1 of a vehicle other than the vehicle 101 can be directly acquired, the execution unit 43 may perform the simulation based on this detection data D1 instead of the usage information of the vehicle 101.

-Third transmission control section 44
The third transmission control unit 44 controls the server-side communication unit 5 when the execution unit 43 performs the simulation. Thereby, the server-side communication unit 5 wirelessly transmits the behavior instruction generated by the execution unit 43 to at least one of the vehicle 101 and the terminal device owned by the traffic participant. By doing this, the occupants of the vehicle 101 (traffic participants) can see the simulation results displayed on the display unit 3 (the display unit of the terminal device) and see what kind of situation they may be placed in in the future. You can find out if there is. As a result, the occupants of the vehicle 101 (traffic participants) can drive (move) with high safety while being aware of the informed situation. This also makes cities and human settlements safer. This will contribute to achieving Goal 11 of the Sustainable Development Goals (SDGs), ``Creating sustainable cities.''

(server and others)
The determination unit 12 of the vehicle 101 determines whether or not an error greater than a threshold has occurred between the future state based on the behavior instruction received from the server 102 and the current state based on the detection data D1 acquired by the acquisition unit 11. , when the learned model 61 is configured to determine whether or not to transmit the usage information, the learned model 61 uses relatively fewer types of detection data D1 (for example, update information D13 and map information D14) than those described above. The prediction data D2 may be generated based on the prediction data D2. Further, in that case, the learned model 61 may be configured to predict the state of the vehicle 101 in the relatively near future. The relatively near future refers to, for example, a time a second predetermined time after the time when the server 102 acquires the detection data D1. The second predetermined time can be set to an arbitrary length that exceeds the acquisition cycle of the detection data D1 but does not exceed the first predetermined time (for example, within a range of 350 to 450 msec). Further, the second predetermined time may include the time from when the server 102 transmits the prediction data D2 until the vehicle 101 receives it. In this case, the state of the vehicle 101 in the near future output by the learned model 61 becomes a predicted value of the detection data D1 after a second predetermined time corresponding to the input detection data D1.

[Digital Twin System 100 Others]
The digital twin system 100 includes a vehicle that has a function to output detection data D1 but does not have a function to obtain simulation results, and a vehicle that does not have a function to output detection data D1 but obtains simulation results. The vehicle may include at least one of the following vehicles.

[Operation of digital twin system 100]
When a predetermined event occurs in the vehicle 101 constituting the digital twin system 100, the vehicle 101 transmits detection data D1 as shown in FIG. 6 (T11). The predetermined events include setting route information in the navigation system, various sensors S included in the vehicle 101 generating detection data D1, and the like. In the case of a conventional vehicle, communication with the server 102 is repeated thereafter (T12, T13, etc.), but after transmitting the detection data D1 at time T11, the vehicle 101 returns to the predicted future (T1t). The detection data D1 is not transmitted until it is determined that the detection data D1 has arrived.

The server 102 that has received the detection data D1 uses the learned model 61 to output prediction data D2 indicating the state of the vehicle 101 in the future (T2t). Then, the server 102 performs a simulation based on the output prediction data D2, and generates a behavior instruction that instructs the state of the vehicle 101 in the future (at time T1t). Then, the server 102 transmits the generated behavior instruction to the vehicle 101.

The vehicle 101 that has received the behavior instruction determines whether or not it is possible to travel based on the behavior instruction, outputs the behavior instruction, etc. Furthermore, when the vehicle 101 reaches the previously predicted future (T1t) (here, at the time (T21) immediately before the previously predicted future (T1t)), the vehicle 101 again transmits the detection data D1 to the server. 102 (T21). After transmitting the detection data D1 at time T21, the vehicle 101 does not transmit the detection data D1 until it determines that it will reach the predicted future (T2t).

The server 102, which has received the detection data D1, uses the learned model 61 to output prediction data D2 indicating the state of the vehicle 101 in the future (T2t). Then, the server 102 performs a simulation based on the output prediction data D2, and generates a behavior instruction that instructs the state of the vehicle 101 in the future (at time T2t). Then, the server 102 transmits the generated behavior instruction to the vehicle 101. Thereafter, the vehicle 101 and the server 102 repeat the operations described above every time they reach the previously predicted future (T31...). As a result, while conventionally data was transmitted periodically as shown in the upper part of FIG. The detection data D1 will be transmitted at longer intervals.

The determination unit 12 of the vehicle 101 determines whether an error greater than a threshold value has occurred between the future state (for example, the time after the first predetermined time from the time when the server 102 acquired the above-mentioned detection data D1) and the current state. If configured to determine whether or not the digital twin system 100 further operates as follows. The vehicle 101 that has received the behavior instruction acquires detection data D1 every time a predetermined period of time has elapsed. Then, it is repeatedly determined whether an error equal to or greater than a threshold value has occurred between the future state based on the behavior instruction received from the server 102 and the current state based on the detection data D1 acquired by the acquisition unit 11. If it is determined that an error greater than the threshold value has occurred, the vehicle 101 transmits the detection data D1 to the server 102 without waiting for the arrival of the next transmission time (T21) of the detection data D1.

In this case, the server 102 that has received the detection data D1 uses the learned model 61 to output the predicted data D2 after the second predetermined time (350 to 450 msec) described above. Then, the server 102 performs a simulation based on the output prediction data D2, and generates a behavior instruction that instructs the state of the vehicle 101 after a second predetermined period of time. Then, the server 102 transmits the generated behavior instruction to the vehicle 101.

The vehicle 101 that has received the behavior instruction determines whether or not it is possible to travel based on the behavior instruction, outputs the behavior instruction, etc. The vehicle 101 that has received the behavior instruction determines whether an error greater than a threshold value has occurred between the future state based on the behavior instruction received from the server 102 and the current state based on the detection data D1 acquired by the acquisition unit 11. It is repeatedly determined whether or not. Then, when the state in which it is determined that no error greater than the threshold value has occurred continues for a second predetermined period of time, the vehicle 101 returns the transmission operation of the detection data D1 to normal. That is, the vehicle 101 returns to the operation of transmitting the detection data D1 to the server 102 when the second predetermined time has elapsed after receiving the behavior instruction. Then, the predicted data D2 outputted by the server 102 that received the detected data D1 also returns to the predicted data D2 after the first predetermined time (for example, 600 msec) described above. On the other hand, if it is determined that an error equal to or greater than the threshold has occurred between the future state and the current state, the vehicle 101 detects the data without waiting for the arrival of the transmission time of the next detection data D1 (after the second predetermined time). Data D1 is sent to the server 102.

The server 102, which has received the detection data D1, uses the learned model 61 to output predicted data D2 after a third predetermined time period (for example, within a range of 200 to 300 msec) that is shorter than the second predetermined time period. Note that the third predetermined time may be longer than the acquisition cycle of the detection data D1. Further, the third predetermined time may include the time from when the server 102 transmits the prediction data D2 until the vehicle 101 receives it. Then, the server 102 performs a simulation based on the output prediction data D2, and generates a behavior instruction that instructs the state of the vehicle 101 after a third predetermined period of time. Then, the server 102 transmits the generated behavior instruction to the vehicle 101.

The vehicle 101 that has received the behavior instruction determines whether an error greater than a threshold value has occurred between the future state based on the behavior instruction received from the server 102 and the current state based on the detection data D1 acquired by the acquisition unit 11. is repeatedly determined. Then, when the state in which it is determined that no error greater than the threshold value has occurred continues for a third predetermined period of time, the vehicle 101 returns to the normal transmission operation of the detection data D1 after receiving the next behavior instruction. Then, the predicted data D2 outputted by the server 102 that received the detected data D1 also returns to the predicted data D2 after the second predetermined time (350 to 450 msec) described above. On the other hand, if it is determined that an error equal to or greater than the threshold has occurred between the future state and the current state, the vehicle 101 detects the data without waiting for the transmission time of the next detection data D1 (after the third predetermined time). Data D1 is sent to the server 102. Thereafter, the vehicle 101 and the server 102 repeat the operations described above.

By doing this, when the vehicle 101 is in an expected state, the amount of communication between the vehicle 101 and the server 102 is reduced, and when an unexpected situation occurs in the vehicle 101, or when an unexpected situation occurs, the amount of communication between the vehicle 101 and the server 102 is reduced. If this continues, the amount of communication will increase somewhat, but the server 102 will be able to grasp the status of the vehicle 101 in more real time.

[Effect]
According to the vehicle 101 according to the present embodiment described above, since the detection data D1 is transmitted to the server 102 and the future state of the vehicle 101 is predicted on the server 102 side, the server 102 has a sufficient amount of detection data. Based on D1, it is possible to predict the distant future with higher accuracy and generate accurate behavior instructions. Furthermore, by doing so, there is no need to transmit data to the server 102 until the predicted future is reached, so the amount of communication between the vehicle 101 and the server 102 can be reduced. As a result, in the digital twin system 100 that includes the vehicle 101 and the server 102 that generates the digital twin of the vehicle 101, the amount of communication between the vehicle 101 and the server 102 can be reduced without reducing the accuracy of the digital twin. be able to.

<Other embodiments>
Part or all of the functions of each of the control blocks described above can also be realized by a logic circuit. For example, an integrated circuit in which a logic circuit functioning as each of the control blocks described above is formed is also included in the scope of the present invention. In addition to this, it is also possible to realize the functions of each of the control blocks described above using, for example, a quantum computer.

[Example of implementation using software]
The functions of the vehicle 101 and the server 102 (hereinafter referred to as "devices, etc.") are programs for making a computer function as the devices, and each control block of the device (in particular, the vehicle-side control unit 1, the server Each unit included in the side control unit 4) can be realized by a program for causing a computer to function. In this case, the device includes a computer having at least one control device (for example, a processor) and at least one storage device (for example, a memory) as hardware for executing the program. By executing the above program using this control device and storage device, each function described in each of the above embodiments is realized. The above program may be recorded on one or more computer-readable recording media instead of temporary. This recording medium may or may not be included in the above device. In the latter case, the program may be supplied to the device via any transmission medium, wired or wireless.

In addition, according to each aspect of the present invention described above, by achieving the above-mentioned effects, it contributes to the achievement of Goal 9 of the Sustainable Development Goals (SDGs), "Create a foundation for industrial and technological innovation." can. Note that the present invention is not limited to the embodiments described above, and can be modified in various ways within the scope of the claims, and can be obtained by appropriately combining technical means disclosed in different embodiments. The embodiments are also included in the technical scope of the present invention.

100 Digital Twin System 101 Vehicle 1 Vehicle-side control section 11 Acquisition section 12 Judgment section 13 First transmission control section 14 First reception control section 15 Judgment section 16 Second transmission control section 17 Output control section 2 Vehicle-side communication section 3 Display section 102 Server 4 Server-side control unit 41 Second reception control unit 42 Prediction unit 43 Execution unit 44 Third transmission control unit 5 Server-side communication unit 6 Storage unit 61 Learned model

Claims (8)

multiple sensors and
an acquisition unit that acquires detection data from each of the plurality of sensors;
At least a part of the detection data acquired by the acquisition unit is transmitted to a server that predicts the future state of the vehicle based on the input of the detection data using a trained model constructed by machine learning, and the server predicts the future state of the vehicle. a communication unit that receives behavior instructions based on the
In the vehicle, the communication unit transmits the next detection data based on a determination result indicating whether or not to transmit the next detection data after transmitting the detection data. comprising a determination unit that determines whether or not running is possible based on the behavior instruction received by the communication unit;
The communication unit transmits the detection data when the determination unit determines that running based on the behavior instruction is not possible.
The vehicle according to claim 1. 3. The vehicle according to claim 1, further comprising a driving control section that automatically controls the operation of at least a portion of the vehicle based on the behavior instruction received by the communication section, or an output control section that controls output of the behavior instruction. vehicle. The vehicle according to any one of claims 1 to 3, further comprising a determination unit that determines whether or not the detection data can be transmitted. The determination unit determines whether or not the time arrives after a predetermined time as whether or not the detection data can be transmitted;
The vehicle according to claim 4, wherein the communication unit transmits the detection data to the server when the determination unit determines that a time after the predetermined time has arrived. a communication unit that receives detection data output from the plurality of sensors from a vehicle including a plurality of sensors and an acquisition unit that acquires detection data from the plurality of sensors, respectively;
a prediction unit that predicts the future state of the vehicle based on the detection data received by the communication unit using a learned model constructed by machine learning, and outputs the prediction result as prediction data;
an execution unit that performs a simulation regarding traffic conditions based on the received detection data and traffic data of traffic participants;
The communication unit is a server, wherein after receiving the detection data, the communication unit receives the next detection data according to a result of determining whether or not the vehicle can transmit the next detection data. The server according to claim 6, wherein the prediction unit requests the vehicle to transmit detection data for new prediction immediately before a time after a predetermined time arrives. The server according to claim 6 or 7, wherein the communication unit transmits the result of the simulation performed by the execution unit to at least one of a terminal device owned by the vehicle and the traffic participant.

Images